Research map / Image generation / Autoregressive and latent generation
ScanDMM: A Deep Markov Model of Scanpath Prediction for 360deg Images
Xiangjie Sui, Yuming Fang, Hanwei Zhu, Shiqi Wang, Zhou Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add state initialization conditioned on the scanpath starting point (ablation) | ScanDMM · Sitzmann | LEV: 45.949 -> 44.966 | Improved |
| Add scene semantics to the transition function (ablation) | ScanDMM · Sitzmann | LEV: 46.800 -> 44.966 | Improved |
| Add state initialization conditioned on the scanpath starting point (ablation) | ScanDMM · Salient360! | LEV: 38.340 -> 37.272 | Improved |
| Add scene semantics to the transition function (ablation) | ScanDMM · Salient360! | LEV: 39.194 -> 37.272 | Improved |
| Add both starting-point initialization and scene semantics (ablation) | ScanDMM · Sitzmann | LEV: 46.489 -> 44.966 | Improved |
| Add both starting-point initialization and scene semantics (ablation) | ScanDMM · Salient360! | LEV: 38.958 -> 37.272 | Improved |
| Use ScanDMM instead of the strongest baseline on each dataset | 360-degree scanpath prediction · Sitzmann, Salient360!, AOI, JUFE | REC: 3.241 -> 3.475; 3.099 -> 3.576; 3.750 -> 4.024; 3.075 -> 4.329 | Improved |
| Use ScanDMM-generated scanpaths rather than direct 2D IQA on equirectangular projections | 360-degree image quality assessment · JUFE | CC: 0.156 -> 0.563; 0.148 -> 0.519; 0.163 -> 0.597; 0.162 -> 0.662; 0.160 -> 0.635 | Improved |
Later papers that compared with this method
2 later accepted papers report ScanDMM as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- SalDiff-DTM: A Novel Dual-Temporal Modulated Diffusion Model for Omnidirectional Images Scanpath Prediction (AAAI 2026 · better on 8 of 9 shared results)
- Assessor360: Multi-sequence Network for Blind Omnidirectional Image Quality Assessment (NeurIPS 2023 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ST-DiffEye: Diffusion-based Continuous Gaze Generation via Joint Scanpath-Trajectory Modeling (NeurIPS 2026 · similarity 0.85)
- DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images (NeurIPS 2025 · similarity 0.84)
- Unifying Top-down and Bottom-up Scanpath Prediction Using Transformers (CVPR 2024 · similarity 0.84)
- Few-shot Personalized Scanpath Prediction (CVPR 2025 · similarity 0.84)
- Beyond Average: Individualized Visual Scanpath Prediction (CVPR 2024 · similarity 0.84)
- Attention (as Discrete-Time Markov) Chains (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
